Mutahir Ahmed

I’m an AI Developer specialising in RAG pipelines, multi-agent systems, and LLM-powered tools. I’ve built and deployed production-ready applications, and I bring client-facing experience—translating business requirements into working AI solutions for both technical and non-technical stakeholders.

My research background includes a published deep learning paper (ICGCET 2021), where I built a CNN + SSD pipeline to detect and classify Egyptian hieroglyphs. That work directly informs my current “Hieroglyph Explorer” RAG application, reflecting how I turn research into usable product experiences.

Professionally, I’ve built 7 production AI applications on Streamlit Cloud, including a CrewAI customer support system with FAISS-backed RAG, visible multi-agent orchestration, and real-time per-agent output. I also created a multi-document intelligence agent that supports cross-document semantic search with source-cited answers and query enrichment for multi-turn conversations.

Beyond build-and-deploy, I enjoy designing agent behaviors and reliability loops—using LangChain AgentExecutor with ReAct reasoning and tools like Tavily, Groq, and Groq’s OpenAI-compatible APIs. I’m especially interested in roles where I can keep shipping high-impact LLM systems end-to-end.